From: Vitaly Wool <vitalywool@xxxxxxxxx> Subject: mm/z3fold.c: claim page in the beginning of free There's a really hard to reproduce race in z3fold between z3fold_free() and z3fold_reclaim_page(). z3fold_reclaim_page() can claim the page after z3fold_free() has checked if the page was claimed and z3fold_free() will then schedule this page for compaction which may in turn lead to random page faults (since that page would have been reclaimed by then). Fix that by claiming page in the beginning of z3fold_free() and not forgetting to clear the claim in the end. [vitalywool@xxxxxxxxx: v2] Link: http://lkml.kernel.org/r/20190928113456.152742cf@bigdell Link: http://lkml.kernel.org/r/20190926104844.4f0c6efa1366b8f5741eaba9@xxxxxxxxx Signed-off-by: Vitaly Wool <vitalywool@xxxxxxxxx> Reported-by: Markus Linnala <markus.linnala@xxxxxxxxx> Cc: Dan Streetman <ddstreet@xxxxxxxx> Cc: Vlastimil Babka <vbabka@xxxxxxx> Cc: Henry Burns <henrywolfeburns@xxxxxxxxx> Cc: Shakeel Butt <shakeelb@xxxxxxxxxx> Cc: Markus Linnala <markus.linnala@xxxxxxxxx> Cc: <stable@xxxxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/z3fold.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) --- a/mm/z3fold.c~z3fold-claim-page-in-the-beginning-of-free +++ a/mm/z3fold.c @@ -998,9 +998,11 @@ static void z3fold_free(struct z3fold_po struct z3fold_header *zhdr; struct page *page; enum buddy bud; + bool page_claimed; zhdr = handle_to_z3fold_header(handle); page = virt_to_page(zhdr); + page_claimed = test_and_set_bit(PAGE_CLAIMED, &page->private); if (test_bit(PAGE_HEADLESS, &page->private)) { /* if a headless page is under reclaim, just leave. @@ -1008,7 +1010,7 @@ static void z3fold_free(struct z3fold_po * has not been set before, we release this page * immediately so we don't care about its value any more. */ - if (!test_and_set_bit(PAGE_CLAIMED, &page->private)) { + if (!page_claimed) { spin_lock(&pool->lock); list_del(&page->lru); spin_unlock(&pool->lock); @@ -1044,13 +1046,15 @@ static void z3fold_free(struct z3fold_po atomic64_dec(&pool->pages_nr); return; } - if (test_bit(PAGE_CLAIMED, &page->private)) { + if (page_claimed) { + /* the page has not been claimed by us */ z3fold_page_unlock(zhdr); return; } if (unlikely(PageIsolated(page)) || test_and_set_bit(NEEDS_COMPACTING, &page->private)) { z3fold_page_unlock(zhdr); + clear_bit(PAGE_CLAIMED, &page->private); return; } if (zhdr->cpu < 0 || !cpu_online(zhdr->cpu)) { @@ -1060,10 +1064,12 @@ static void z3fold_free(struct z3fold_po zhdr->cpu = -1; kref_get(&zhdr->refcount); do_compact_page(zhdr, true); + clear_bit(PAGE_CLAIMED, &page->private); return; } kref_get(&zhdr->refcount); queue_work_on(zhdr->cpu, pool->compact_wq, &zhdr->work); + clear_bit(PAGE_CLAIMED, &page->private); z3fold_page_unlock(zhdr); } _
